Is There a Transition-State Imbalance in Malononitrile Anion Forming

that the deprotonation of malononitrile derivatives by carboxylate ions should also have an imbalanced transition state. The fact that none has been o...
0 downloads 0 Views 982KB Size
J. Org. Chem. 1989,54, 2878-2885

2878

Is There a Transition-State Imbalance in Malononitrile Anion Forming Reactions? Kinetics of Piperidine and Morpholine Addition to Substituted Benzylidenemalononitriles in Various Me2SO-Water Mixtures' Claude F. Bernasconi* and Robert B. Killion, Jr. Thimann Laboratories of the University of California, Santa Cruz, California 95064

Received February 13, 1989 Piperidine and morpholine add to substituted benzylidenemalononitriles (Z-C6H4CH=C(CN)2)to form a zwitterionic adduct, Z-C6H4CH(R2NH+)C(CN)2(T*), which is in rapid acid-base equilibrium with the anionic adduct, Z-C6H&H(R2N)C(CN),- (T). Rate constants for amine addition ( k , ) were determined by direct rate measurementswhile equilibrium constants for addition (Kl)as well as pK,* values of the zwitterionswere obtained spectrophotometrically. The bulk of the measurementswas carried out in 5090Me&O-50% water with piperidine, while a smaller number of experiments were performed with morpholine, and with both amines in water and in 70% MezSO-30% water. The reactions show the typical behavior of a carbanion-forming process in which the carbanion derives a good part of its stabilization from polar effects while resonance effects play a more modest role. This behavior includes a high intrinsic rate constant (k, = k when K = l),a small transition-state imbalance, and a relatively small solvent effect on the intrinsic rate constant. The observation of an imbalance suggests that the deprotonation of malononitrile derivatives by carboxylate ions should also have an imbalanced transition state. The fact that none has been observed is attributed to a solvation effect of the carboxylic acid, which enhances the Br~nstedPB value, as recently suggested by Murray and Jencks. The 4-Me2Nsubstituent leads to strong resonance stabilization of the olefin ag indicated by a low K1 value. Contrary to expectation of a lowered intrinsic constant, this resonance stabiliiati0.n has little effect on k@ This suggests the operation of a compensating factor which increases ko and which can be understood as an attenuation of the reduction in ko caused by late development of resonance at the carbanionic center of the adduct. In our recent work on structure-reactivity relationships in carbanion-forming reactions we have been interested in similarities between nucleophilic additions to activated olefins, eq 1, and proton transfers from activated carbon, eq 2. Both reactions lead to the formation of a carbanion X CH=CXY

z

+

Nu'

S k-1

NUZ+ 1

of nearly identical structure except that in eq 1 the hydrogen attached to the central carbon is replaced by the group ZC6H4CH(NuZ+').It is therefore not surprising that both reactions show a qualitatively similar dependence of the intrinsic rate constant (k, = k, = k-, when K1 = 1)on the activating groups X and Y. This dependence is such that ko decreases with increasing resonance stabilization of the carbanion, Le., (CN), > (COO),C(CH,), > CN(C6Hd-4-No2) > CN(C6H,-2,4-(NO2)2) > H(N02) > Ph(NO,).,,, The observed trend can be understood in terms of transition-state structures in which resonance development and concomitant solvation of the negative charge lag behind bond formation. As a consequence of the principle of nonperfect synchronization (PNS),2-4this lag leads to a reduction in ko relative to a system in which there is no such resonance. This reduction becomes larger with increasing resonance in the carbanion which explains the observed trend in ko with XY. (1) This is Part 24 in the series Nucleophilic Addition to Olefms. Part 23: Bernasconi, C. F.;Killion, R. B., Jr. J. Am. Chem. SOC.1988, 110, 7506. (2) Bernasconi, C. F.Acc. Chem. Res. 1987,20, 301. (3) Bernasconi, C. F.In Nucleophilicity; Harris, J. M., McManus, S. P., Eds.; Advances in Chemistry 215; American Chemical Society: Washington, DC, 1987; p 137. (4) Bernasconi, C. F. Tetrahedron 1985, 41, 3219.

0022-326318911954-2878$01.50/0

Another manifestation of the lag in resonance development behind bond formation is the disparity or imbalance in the Bransted coefficients of these reactions. In the proton transfers this imbalance is measured by I = acH5 - pB5 > 0, in the nucleophilic addition reactions by I = a,," - pnB,," > 0. The imbalances are large for reactions that lead to carbanions with strong resonance stabilization and small when resonance plays only a minor role. Quantitative comparisons of the effect of XY on intrinsic rate constants and imbalances show significant differences between reactions 1and 2 though. ko for the nucleophilic additions is less sensitive to XY than ko for proton transfers: a plot of log ko (eq 1)vs log k, (eq 2) gives a fairly good linear correlation with a slope of =0.45., Similarly, the imbalances observed for the nucleophilic additions are generally smaller than for the corresponding proton transfers. For example, I = 0.67 - 0.37 = 0.30 for PhCH=C(Ar)N02 + R2NH7while I = 1.29 - 0.55 = 0.74 for ArCH2N02 R,NH;* or I = 0.22 - 0.07 = 0.15 for ArCH=C(COO),C(CH& + R2NH9while I = 0.76 - 0.44 = 0.32 for ArCH,CH(COMe)COOEt + RCOO-.1° There is one proton-transfer reaction, the deprotonation of malononitrile derivatives, that seems to have an abnormally small imbalance. For example, for the reaction of ArCH2CH(CN)2with RCOO-, aCH = /3B = 1.0 and thus I = 0.loa Similar resultslobwere reported for the detritiation of tert-butylmalononitrile by RCOO- where /3B = 0.98 while an aCH = 1.06 f 0.05 was calculated from the rates of detritiation of tert-butylmalononitrile and malononitrile by formate ion. A somewhat lower PB = 0.8 was found for the deprotonation of tert-butylmalononitrile by primary amines'& but in the absence of an experimental CYCH value

+

(5) CXCH= d log k,/d log KacH,e.g., when pK,"" is varied by varying Z in Z-C6H4CHXY;pB = d log k1/dpKaBH. (6) an," = d log kl/d log K1,e.g., by varying Z in eq 1; p,", = d log kl/d log K , by varying Nu. (7) Bernasconi, C. F.; Renfrow, R. A. J. O g . Chem. 1987, 52, 3035. (8) Bordwell, F. G.; Boyle, W. J., Jr. J. Am. Chem. SOC.1972,94,3907. (9) Bernasconi, C. F.;Panda, M. J. Org. Chem. 1987, 52, 3042. (10)(a) Bell, R. P.; Grainger, S. J. Chem. SOC.,Perkin Trans.2 1976, 1367. (b)Hibbert, F.;Long, F. A.; Welters, E. A. J.Am. Chem. Soc. 1971, 93, 2829. (c) Pratt, R. F.; Bruice, T. C. J. Org. Chem. 1972, 37, 3563.

0 1989 American Chemical Society

J. Org. Chem., Vol. 54,No. 12, 1989 2879

Kinetics of Piperidine and Morpholine Addition Scheme I

500

I

400

S

300 L

I

In

L

'F

T-

T i

"$+

200

100

2

Y

R2N TO

01 0

it is not clear whether these results imply a larger imbalance for the amine reactions. We shall return to this point in the Discussion. Even if one assumes a rather modest resonance effect in the malononitrile type a modest imbalance that is larger than zero would seem to fit better with the pattern established by the other examples. It occurred to us that one way to demonstrate that the above zero imbalance is indeed abnormal and probably caused by special factors coming into play in the malononitrile deprotonation would be to show that the imbalance in the nucleophilic addition to a dicyano activated olefin is larger than zero. To this end we have studied the reaction of piperidine and morpholine with substituted benzylidenemalononitriles, eq. 3.

L

L

I

R*NH+

(3)

A second motivation for our study was to examine the effect of n-donor substituents (Z = 4-Me2N,4-Me0) on the intrinsic rate constant of reaction 3. Two interaction mechanisms of the ?r-donorshave recently been identified which can affect kP9J4 The first one leads to a decrease in ko and can be visualized as a loss of the resonance stabilization of the olefin running ahead of bond formation. The second leads to an increase in ko and can be understood as an attenuation of the decrease in ko caused by the late development of the resonance effect at the carbanionic center of the product ion. For systems where carbanion resonance is strong the second factor dominates, leading to a net increase in k0,I4 while for systems where the resonance effect in the adduct is weak one might expect that the first interaction mechanism may override the second one, resulting in a net decrease in k,,. Among the reactions studied in this context, reaction 3 involves the smallest degree of resonance stabilization in the adduct and hence should provide a good test case for the above prediction. A third point of interest is the effect of increasing the Me&O content of the solvent on the intrinsic rate constant of the reaction. We therefore determined ko for piperidine (11)Bell, R. P. The Proton in Chemistry; Cornel1 University Press: lthaca, NY, 1973; p 105. (12)Hine, J. Adv. Phys. Org. Chem. 1977,15, 1. (13)(a) Bordwell, F. G.; Bares, J. E.; Bartmess, J. E.; McCollum, G. J.; Van der Puy, M.; Vanier, N. R.; Matthews, W. S.J. Org. Chem. 1977, 41,321. (b)Algrim, D.;Bares, J. E.; Branca, J. C.; Bordwell, F. G. Ibid. 1978,43,5024. (14)Bernaaconi, C. F.; Renfrow, R. A.; Tia, P. R. J . Am. Chem. SOC. 1986,108,4541.

A

'

I

0.002

'

"

0.004

I

0.006

"

0.008

p-Me0 "

0.010

[Pip], M

Figure 1. Kinetics of piperidine addition to substituted benzylidenemalononitriles in water at 20 "C. The slopes provide kl according to eq 5.

and morpholine addition to benzylidenemalononitrilein water and in 70% Me2S0, which complements an earlier investigation in 50% Me2S0.15

Results General Features. The reactions of piperidine and morpholine with benzylidenemalononitrile(Z = H) have been previously studied in 50% Me2SO-50% water, p = 0.5 M (KCl), T = 20 OC.15 Two kinetic processes were observed. The first process, shown in Scheme I, was attributed to formation of a zwitterionic addition complex (T*) and subsequent rapid deprotonation to form the anionic complex (T) With . morpholine, a To species was also formed in equilibrium with T* and T.The second process was shown to represent a multi-step breakdown of TO into benzaldehyde and malononitrile. In the present study, the rate and equilibrium constants for the following reactions were determined: piperidine and morpholine with variously substituted benzylidenemalononitrile (BMN-Z) where Z = 4-Me0, H, 3-C1 and 4-NO2, in water; piperidine with BMN-Z where Z = 4Me2N, 4-Me0,4-Me, H, 4-C1,4-Br, 3-C1,4-CN, 4-N02, in 50% Me2SO-50% water (v/v); piperidine and morpholine with BMN-H in 70% Me2SO-30% water (v/v). The temperature was 20 OC in all solvents, the ionic strength 0.5 M in water and 50% Me2S0, 0.25 M in 70% Me2S0, maintained with KC1. Kinetics of Piperidine Addition. In our previous report15 it was shown that with piperidine pKa* >> pK;, implying KO